为什么此代码返回细分错误错误? [英] Why this code returns Segmentation fault error?

问题描述

我实现了以下代码以对矩阵执行filter2D.在我编译程序后，它返回Segmentation Fault错误.在此程序中，我想在程序中分配一个输入数组(不想加载图像).然后执行一个在程序内部分配的过滤器，以在运行时测量不同矩阵和内核的filter2D函数的时间.

I have implemented the following code to perform filter2D on a matrix. After I compiled the program it returns Segmentation fault error. In this program, I want to assign an input array in the program (don't want to load images). then perform a filter which is assigned inside the program to measure the time of the filter2D function for different matrices and kernels at runtime.

#include "opencv2/imgproc/imgproc.hpp"
#include "opencv2/highgui/highgui.hpp"
#include <stdlib.h>
#include <stdio.h>
#include <iostream>

using namespace std;
using namespace cv;

int main(){

    //Picture size for input and output are the same
    int Pxsize = 128;
    int Pysize = Pxsize;

    //Kernel size
    int Kxsize = 3;
    int Kysize = Kxsize;

    //filter arguments
    Point anchor;
    double delta;
    int ddepth;

    //name for out put
    char window_name[32] = "filter2D Demo";


    Mat input[128][128];
    Mat output[128][128];
    Mat kernel[3][3];

    // Initialize arguments for the filter
    anchor = Point( -1, -1 );
    delta = 0;
    ddepth = -1;

    int i,j;
    //assign data between 0 and 255 to the input matrix
    for (i=0; i<Pxsize; i++)
        for (j=0; j<Pysize; j++)
            input[i][j]=(i*j)%255;

    //assign data to the kernel
    //assign data between 0 and 255 to the input matrix
    for (i=0; i<Kxsize; i++)
        for (j=0; j<Kysize; j++)
            kernel[i][j]=1;

    //the problem is here:
    filter2D((InputArray) input,  (OutputArray) output, ddepth , ( InputArray) kernel, anchor, delta, BORDER_DEFAULT );
    namedWindow( window_name, CV_WINDOW_AUTOSIZE );
    imshow( window_name, (OutputArray) output );


    return 0;
}

输出为:

Segmentation fault


------------------
(program exited with code: 139)
Press return to continue

我在Linux Mint中的 gcc 6.2.0 编译器中使用了此命令

I used this command for gcc 6.2.0 compiler in Linux mint

g++ -Wall -Wextra -Werror -pedantic -I/usr/local/include/opencv -I/usr/local/include/opencv2 -L/usr/local/lib/ -g -o "opencv"  "opencv.cpp"  -lopencv_core -lopencv_imgproc -lopencv_highgui -lopencv_ml -lopencv_video -lopencv_features2d -lopencv_calib3d -lopencv_objdetect -lopencv_contrib -lopencv_legacy -lopencv_stitching

编译成功完成

预先感谢

推荐答案

根据评论，发现了问题！这是因为将 Mat 对象转换为 filter2D 函数中的 InputArray 和 OutputArray .我将实现更改为以下程序，并且可以正常运行.

According to the comments, the problem has been found! It is because of casting the Mat object to InputArray and OutputArray in filter2D function. I changed the implementation to the following program and it works.

#include "opencv2/imgproc/imgproc.hpp"
#include "opencv2/highgui/highgui.hpp"
#include <stdlib.h>
#include <stdio.h>
#include <iostream>
#include <time.h>

//Picture size for input and output are the same
#define MAX1 4096

#define Pxsize MAX1
#define Pysize Pxsize

//Kernel size
#define Kxsize 9
#define Kysize Kxsize
//number of iterations
#define NUM_LOOP 10000

using namespace std;
using namespace cv;

unsigned short int A[Pxsize][Pysize];
unsigned short int B[Pxsize][Pysize];
unsigned short int K[Kxsize][Kysize];

// measuring the time
double pTime = 0, mTime = 10; // to indicate how much program time is (pTime) and how much i want to wait (mTime) 
struct timespec start;


int main(){
    //singleCore
    struct timespec tStart, tEnd;//used to record the processiing time
    double tTotal , tBest=10000;//minimum of toltal time will asign to the best time
    int w=0;

    //filter arguments
    Point anchor;
    double delta;
    int ddepth;

    //assign data between 0 and 255 to the input matrix
    int i,j;
    for (i=0; i<Pxsize; i++)
        for (j=0; j<Pysize; j++)
            A[i][j] = (i+j)%255;

    //assign data to the kernel
    //assign data between 0 and 255 to the input matrix
    for (i=0; i<Kxsize; i++)
        for (j=0; j<Kysize; j++)
            K[i][j]=1;

    cv::Mat input = cv::Mat(Pxsize, Pysize, CV_8UC1, A);
    cv::Mat kernel = cv::Mat::ones(Kxsize, Kysize, CV_8UC1);

    //name for out put

    // Initialize arguments for the filter
    anchor = Point( -1, -1 );
    delta = 0;
    ddepth = -1;

    do{// this loop repeat the body to record the best time
        clock_gettime(CLOCK_MONOTONIC,&tStart);

        //the problem is here:
        filter2D( input,   output, ddepth ,  kernel, anchor, delta, 1 );

        clock_gettime(CLOCK_MONOTONIC,&tEnd);
        tTotal = (tEnd.tv_sec - tStart.tv_sec);
        tTotal += (tEnd.tv_nsec - tStart.tv_nsec) / 1000000000.0;

        if(tTotal<tBest)
            tBest=tTotal;
        pTime += tTotal;
    } while(w++ < NUM_LOOP && pTime < mTime);

    //cout<<" "<< input<<endl;
    //cout<<" "<< output<<endl;
    printf("\nThe best time: %f sec in %d repetition for %dX%d matrix for %d, %d kernel \n",tBest,w, Pxsize, Pysize, Kxsize, Kysize);

    return 0;
}

这篇关于为什么此代码返回细分错误错误?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！